Divergent network connectivity changes in behavioural variant frontotemporal dementia and Alzheimer's disease

Abstract

Resting-state or intrinsic connectivity network functional magnetic resonance imaging provides a new tool for mapping large-scale neural network function and dysfunction. Recently, we showed that behavioural variant frontotemporal dementia and Alzheimer's disease cause atrophy within two major networks, an anterior 'Salience Network' (atrophied in behavioural variant frontotemporal dementia) and a posterior 'Default Mode Network' (atrophied in Alzheimer's disease). These networks exhibit an anti-correlated relationship with each other in the healthy brain. The two diseases also feature divergent symptom-deficit profiles, with behavioural variant frontotemporal dementia undermining social-emotional function and preserving or enhancing visuospatial skills, and Alzheimer's disease showing the inverse pattern. We hypothesized that these disorders would exert opposing connectivity effects within the Salience Network (disrupted in behavioural variant frontotemporal dementia but enhanced in Alzheimer's disease) and the Default Mode Network (disrupted in Alzheimer's disease but enhanced in behavioural variant frontotemporal dementia). With task-free functional magnetic resonance imaging, we tested these ideas in behavioural variant frontotemporal dementia, Alzheimer's disease and healthy age-matched controls (n = 12 per group), using independent component analyses to generate group-level network contrasts. As predicted, behavioural variant frontotemporal dementia attenuated Salience Network connectivity, most notably in frontoinsular, cingulate, striatal, thalamic and brainstem nodes, but enhanced connectivity within the Default Mode Network. Alzheimer's disease, in contrast, reduced Default Mode Network connectivity to posterior hippocampus, medial cingulo-parieto-occipital regions and the dorsal raphe nucleus, but intensified Salience Network connectivity. Specific regions of connectivity disruption within each targeted network predicted intrinsic connectivity enhancement within the reciprocal network. In behavioural variant frontotemporal dementia, clinical severity correlated with loss of right frontoinsular Salience Network connectivity and with biparietal Default Mode Network connectivity enhancement. Based on these results, we explored whether a combined index of Salience Network and Default Mode Network connectivity might discriminate between the three groups. Linear discriminant analysis achieved 92% clinical classification accuracy, including 100% separation of behavioural variant frontotemporal dementia and Alzheimer's disease. Patients whose clinical diagnoses were supported by molecular imaging, genetics, or pathology showed 100% separation using this method, including four diagnostically equivocal 'test' patients not used to train the algorithm. Overall, the findings suggest that behavioural variant frontotemporal dementia and Alzheimer's disease lead to divergent network connectivity patterns, consistent with known reciprocal network interactions and the strength and deficit profiles of the two disorders. Further developed, intrinsic connectivity network signatures may provide simple, inexpensive, and non-invasive biomarkers for dementia differential diagnosis and disease monitoring.

Figure 1

Study design schematic. Preprocessed task-free fMRI data were decomposed using ICA, and Salience Network (SN) and DMN components were identified for each subject by calculating goodness-of-fit to network templates derived from an independent dataset of young healthy controls. Grey matter maps were also derived from structural MRI data of each subject for atrophy correction. Hypotheses regarding between-group connectivity alterations were tested using the linear contrasts shown. SN and DMN scores and the combined SN minus DMN index for each subject were derived and entered into three-class linear discriminant analyses. HC = healthy controls; AD = Alzheimer’s disease; VBM = voxel-based morphometry; LDA = linear discriminant analysis.

Figure 2

BvFTD and Alzheimer’s disease feature divergent Salience Network and DMN dynamics. Group difference maps illustrate clusters of significantly reduced or increased connectivity for each ICN. In the Salience Network (A), patients with bvFTD showed distributed connectivity reductions compared to healthy controls (HC) and patients with Alzheimer’s disease (AD), whereas patients with Alzheimer’s disease showed increased connectivity in anterior cingulate cortex and ventral striatum compared to healthy controls. In the DMN (B), patients with Alzheimer’s disease showed several connectivity impairments compared to healthy controls and patients with bvFTD, whereas patients with bvFTD showed increased left angular gyrus connectivity. Patients with bvFTD and Alzheimer’s disease further showed focal brainstem connectivity disruptions within their ‘released’ network (DMN for bvFTD, Salience Network for Alzheimer’s disease). Results are displayed at a joint height and extent probability threshold of P < 0.05, corrected at the whole brain level. Colour bars represent t-scores, and statistical maps are superimposed on the Montreal Neurological Institute template brain.

Figure 3

Subcortical and brainstem Salience Network connectivity reductions in bvFTD. Statistical maps illustrate the Salience Network contrast of bvFTD < controls. Axial slice insets are provided for selected brainstem regions to clarify cluster locations. blA = basolateral nucleus of amygdala; Hc = habenular complex; PBN = parabrachial nucleus; Pc = paracentral nucleus of thalamus; Pf = parafasicular nucleus of thalamus; RedN = red nucleus; SN = substantia nigra; vMD = ventral mediodorsal nuclei; VSP = ventral striatopallidum; VTA = ventral tegmental area; PAG = periaqueductal grey matter. Presentation details otherwise as in Fig. 2.

Figure 4

ICN enhancements relate to specific connectivity disruptions within the reciprocal networks. (A) Left angular gyrus (L ANG) DMN connectivity was intensified in patients with bvFTD versus healthy controls and the 3 mm radius spherical region of interest centred on its peak showed a significant inverse correlation with SN connectivity in right frontoinsula (R FI), anterior mid-cingulate cortex (aMCC) and the dorsal pons (not shown) across patients with bvFTD and healthy controls (HC). Group-wise scatterplots illustrate the relative contribution of each group to these relationships for each anticorrelated region. (B) Right pregenual anterior cingulate cortex (R pACC) SN connectivity, enhanced in patients with Alzheimer’s disease (AD), showed a significant inverse relationship with DMN connectivity in precuneus (PreCu), ventromedial prefrontal cortex (vmPFC) and bilateral middle temporal gyrus (not shown) across patients with Alzheimer’s disease and healthy controls. Results are atrophy-corrected and thresholded at a joint height and extent probability of P < 0.05, corrected at the whole brain level. Colour bars represent t-scores, and statistical maps are superimposed on the Montreal Neurological Institute template brain.

Figure 5

Salience Network disruption and DMN enhancement correlate with bvFTD clinical severity. In bvFTD, CDR, sum of boxes scores (CDR-SB) showed a significant (A) negative correlation with Salience Network connectivity in right frontoinsula (FI) and (B) positive correlation with DMN connectivity in bilateral angular gyrus (ANG), as well as ventromedial prefrontal cortex (vmPFC) and right intraparietal sulcus (R IPS, not shown). Results are atrophy corrected and thresholded at a joint height and extent probability of P < 0.05, corrected at the whole brain level. Colour bars represent t-scores, and statistical maps are superimposed on the Montreal Neurological Institute template brain.

Figure 6

Group discrimination using Salience Network and DMN metrics. (A) Scatterplot of Salience Network (SN) scores of healthy controls (HC), patients with bvFTD and Alzheimer’s disease (AD) with respect to clinical diagnosis. Each subject’s Salience Network score represents the average z-score across fourteen regions of interest derived from a Salience Network (bvFTD < healthy control) + (Alzheimer’s disease > healthy control) conjunction analysis (Supplementary Table 3). (B) DMN scores were computed in a parallel fashion, using eight regions of interest derived from a DMN (Alzheimer’s disease < healthy control) + (bvFTD > healthy control) conjunction analysis (Supplementary Table 3). (C) The combined Salience Network minus DMN index provided superior group discrimination compared to analysis of either network alone. Dot columns were converted to clouds by random number assignment within each group to improve scatter visualization. Key for colours and shapes is provided at the bottom.